The capability of 1.5-T MR imaging to detect focal defects in articular cartilage was investigated with cadaveric knees with and without intraarticular injection of saline and gadolinium-DTPA (Gd-DTPA). Full-thickness cartilage lesions ranging in diameter from 1 to 5 mm were surgically created in the femoral articular surfaces. Images were acquired with a variety of pulse techniques, slice thicknesses, and interslice gaps as well as one or two signal excitations. Potential intraarticular contrast agents (saline and Gd-DTPA) were tested, and their signal behaviors compared with that of hyaline cartilage. All cartilage defects were occult on T1-weighted and balanced images without Gd-DTPA. The smallest defect identified by using intraarticular saline was 3 mm in diameter and was apparent only on T2-weighted images. Intraarticular Gd-DTPA afforded detection of defects as small as 2 mm, even with short imaging times. Signal-intensity differences between saline and articular cartilage were minimal on T1-weighted images and increased on T2-weighted images; intensity differences were high between Gd-DTPA and articular cartilage on all imaging sequences. These results indicate that intraarticular fluid and appropriate selection of imaging sequences are necessary for delineation of focal defects in articular cartilage. They also show that Gd-DTPA is the optimal contrast agent for this purpose.